Droplet Mobility Manipulation on Porous Media Using Backpressure

G.P., G.K., and A.G.B. have received funding from the Hellenic National Strategic ..... phenomenon in membrane contactors – Causes and prevention J...
2 downloads 0 Views 7MB Size
Download PDF
Article pubs.acs.org/Langmuir

Droplet Mobility Manipulation on Porous Media Using Backpressure N. Vourdas,*,† G. Pashos,‡ G. Kokkoris,‡,§ A. G. Boudouvis,‡ and V. N. Stathopoulos*,† †

School of Technological Applications, Technological Educational Institute of Sterea Ellada, Psachna 34400, Evia, Greece School of Chemical Engineering, National Technical University of Athens, Zografou Campus, Athens 15780, Greece § Institute of Nanoscience & Nanotechnology, NCSR Demokritos, Athens 15341, Greece ‡

ABSTRACT: Wetting phenomena on hydrophobic surfaces are strongly related to the volume and pressure of gas pockets residing at the solid−liquid interface. In this study, we explore the underlying mechanisms of droplet actuation and mobility manipulation when backpressure is applied through a porous medium under a sessile pinned droplet. Reversible transitions between the initially sticky state and the slippery states are thus incited by modulating the backpressure. The sliding angles of deionized (DI) water and ethanol in DI water droplets of various volumes are presented to quantify the effect of the backpressure on the droplet mobility. For a 50 μL water droplet, the sliding angle decreases from 45 to 0° when the backpressure increases to ca. 0.60 bar. Significantly smaller backpressure levels are required for lower surface energy liquids. We shed light on the droplet actuation and movement mechanisms by means of simulations encompassing the momentum conservation and the continuity equations along with the Cahn−Hilliard phase-field equations in a 2D computational domain. The droplet actuation mechanism entails depinning of the receding contact line and movement by means of forward wave propagation reaching the front of the droplet. Eventually, the droplet skips forward. The contact line depinning is also corroborated by analytical calculations based on the governing vertical force balance, properly modified to incorporate the effect of the backpressure.

■

INTRODUCTION Control over the mobility of droplets on surfaces remains an important part of recent research activity, both for understanding the appropriate wetting principles1 and for practical applications including open, closed, and digital microfluidics and related technologies,2−8 chemical processes in droplets,9,10 engineered self-cleaning surfaces,11−14 antiicing coatings,15,16 membrane contactors,17 polymer electrolyte fuel cells,18,19 heat exchangers,20,21 and so forth. In this direction, a variety of approaches have been proposed toward preparing active,22 tunable,23 adaptive,24 and skatinglike25 surfaces, namely, manipulating the mobility of a droplet or a liquid film mainly on the basis of the respective wetting transitions. Pertinent methodologies have been reviewed recently.26−30 The vast majority of the methods for active droplet mobility manipulation entails the use of electrical,31 magnetic,32 thermal,33 vibrational,34−36 mechanical,37 or optical stimuli38 and necessitates the fabrication of microelectrodes, specific stacking layer sequences, or surface texturing, requires structures with flexural elasticity, induces heat dissipation, or introduces foreign particles into the liquid phase, which may provide drawbacks for specific applications. Particular interest has been focused on some active methods interfering directly with the volume and the pressure of the gas pockets residing in the textured solid−liquid interface; such characteristics are crucial in such phenomena.39 The presence of a thin lubricating air film under the liquid has proven to play a major role in the mobility of droplets, e.g., after impingement on rough40 or porous41 surfaces as well as on liquids.42 © XXXX American Chemical Society

Moreover, the pressure difference below and over the droplet has been identified as critical in spontaneous droplet depinning.43 Pertinent direct approaches for tuning the gas pockets have found various applications in the underwater restoration of superhydrophobicity44 through electrochemical gas production and in writing/erasing procedures,45 enabling the storage of binary data, by producing an overpressure or underpressure through an orifice. Alternatively, the gas pockets may also be produced and tuned through direct pressure application from the rear side (backpressure), by ejecting gas through one hole41,46,47 or by introducing gas bubbles by means of a pipet.48−51 The limit of this process is the case of droplets levitated by an air cushion via an air stream flowing under a droplet.27,52−55 In this direction, we have recently developed a method for droplet actuation and mobility manipulation on porous media,56−58 which has found application in open-channel fluidic valving59,60 and in mobility control after impingement.41 Contrary to the droplets levitated by an air cushion exhibiting frictionless mobility, our method provokes only partial depinning of the solid−liquid interface, which is enough to render the droplet mobile and to incite downward movement. The main advantages of this include the absence of moving parts and the circumvention of electrical, magnetic, optical, vibrational, and acoustical stimuli on the surface. The heat dissipation may be not only minimized but also finely Received: March 7, 2016 Revised: April 19, 2016

A

DOI: 10.1021/acs.langmuir.6b00900 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

110 °C for 20 min. SL200 KS from Kino was used to measure the contact angle after this treatment. The apparent contact angles were ca. 122 and 45° for water and for 7.5% ethanol (mole fraction) in water, respectively, measured with 5 μL droplets. Experiments have been also performed on membranes deposited via vapor deposition with an appropriate precursor to deliver grafting of self-assembled perfluorododecyltrichlorosilane monolayers (SAMs). A scanning electron microscopy (SEM) image of the D0.9 = 7.1 μm hydrophobic membrane, i.e., after Teflon coating, is presented in Figure 1.

controlled by controlling the temperature of the gas feed. The introduction of foreign particles/molecules into the working liquid/droplet is completely avoided. It may accommodate various liquids; i.e., it may work on deionized (DI) water to aqueous solutions, as will be evidenced hereafter, and is amenable for integration onto devices for active control. Finally, the response time is ultralow, on the order of milliseconds, and if appropriate microporous structure is selected, its energy demands are significantly lower than for other methods. Even though this method has been proven to be rather versatile in delivering mobility manipulation, as described above, we have not yet provided the underlying working mechanism. The application of the backpressure introduces a variety of changes51 in wetting-related parameters, such as in the advancing and mainly in the receding contact angle and in the liquid−solid contact area. Additional Laplace forces appear, which are asymmetric due to the inclination and deform the droplet shape. The air cushion under the droplet is not fully developed; therefore, Leidenfost dynamics61,62 are not suitable to apply. For this reason, we may not assume that the droplet is supported on an airlike lubricating film;63 i.e., the air pockets do not fully bear the gravitational force exerted by the droplet. On the other hand, the droplet does not attain a Wenzel, i.e., fully wetted, state; therefore, well-established Wenzel to Cassie− Baxter, i.e., partially wetted, transition dynamics are not pertinent. The droplet moves along various Cassie−Baxter states enabled by the application of backpressure, some of which correspond to sticky behavior and some, to slippery behavior. Moreover, it has been proven that the force exerted by each bubble alone cannot account for the downward droplet movement.51 In this work, we shed light on the underlying working mechanisms of droplet actuation and movement induced by backpressure application through a porous medium. The problem is followed by means of simulations encompassing the momentum conservation and the continuity equations along with the Cahn−Hilliard phase-field equations in a 2D domain and by employing the governing vertical force balance accounting for the partial depinning of the receding contact line. This depinning is also corroborated by analytical calculations. The pertaining vertical force balance has been modified to incorporate the effect of the backpressure and to yield the pinned fraction of the contact line. The results from the theoretical approach are compared to the experimental observations, thus providing a reasonable understanding of the respective droplet actuation and movement mechanisms.

■

Figure 1. SEM image of the hydrophobic porous medium surface (αAl2O3, D0.9 = 7.1 μm). Measurement of the average pore size and spacing may not be straightforward for such surfaces. Pores cannot be explicitly defined, and their size depends on the actual wetted fraction area, information that cannot be predicted in advance as in the case of lithographically fabricated structures. A detailed discussion on this may be found in a previously published study.41 In any case, the scale of geometrical characteristics of the surface is much lower compared to the nominal wetted area. For the analytical approach, presented below, pore radius values varied from ca. 0.5 to 1.6 μm, and the contact patch radius, from ca. 0.50 to 1.65 μm. The hydrophobic porous medium was positioned on a perforated Plexiglas sheet under which the gas pressure is controlled using a needle valve and a manometer with an accuracy of ±2 mbar (Kimo, model MP 200HP). A schematic representation of the experimental setup is given in Figure 2. In all cases presented herein, compressed air was used as the pneumatic means for actuation and mobility manipulation.

■

EXPERIMENTAL SECTION

RESULTS AND DISCUSSION In Figure 3, the sliding angle of water droplets of various volumes are presented vs the applied backpressure on the porous surface. The sliding angle (a) is defined as the minimum tilt angle at which the droplet moves systematically downward, at a particular backpressure for each droplet volume. The droplet was first positioned on the porous surface, and then the backpressure was gradually increased to the point where sliding took place. Increasing backpressure incites partial droplet depinning, as will be discussed later, thus decreasing the sliding angle. After this step, the gravitational force, FG, induces a downward movement of the droplet. It is expected that as FG increases, less pronounced depinning is needed to induce movement; i.e., the depinning fraction may be lower. We note here and will discuss later that at this backpressure limit, sliding

Porous media (membranes) were fabricated from α-Al2O3 powder with a particle size distribution in which 90% of the particles had a diameter of less than 7.0 μm, i.e., D0.9 = 7.1 μm. Up to 5 wt % (NH4)2HPO4 was added to Al2O3 powder, and the mixture was uniaxially pressed under 110 MPa to form green disk-shaped membranes with a diameter of 13 mm. A similar procedure was followed to form membranes with a diameter of 32 mm. The final membranes were prepared after sintering the green samples for 3 h at 1000 °C. At this stage, the membranes are superhydrophilic and waterabsorbing. The applicability of this method is not limited to this porous structure. Other porous materials may be used with various optical, structural, and electrical characteristics as well.64,65 To render the initially superhydrophilic and water-absorbing surface hydrophobic, all of the membranes were infiltrated with a 0.5 wt % Teflon solution (poly(4,5-difluoro-2,2-bis(trifluoromethyl)-1,3- dioxole-co-tetrafluoroethylene in Fluorinert FC-770) and then heated to B

DOI: 10.1021/acs.langmuir.6b00900 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

Figure 2. Schematic representation of the experimental setup used for the measurement of the sliding angle (α) variation with backpressure. The sliding angle is the minimum inclination angle at which the droplet moves systematically downward.

Figure 3. Sliding angle (a) vs backpressure. Lines of different DI water droplet volumes are depicted.

takes place in a stick−slip kind of movement; i.e., the droplet does not move downward continuously. The sliding angle of a 50 μL water droplet is 45° when no backpressure is applied and gradually drops to 28 and 10° when the backpressure increases to 0.17 and 0.43 bar, respectively. When the backpressure increases to 0.60 bar, the 50 μL water droplet exhibits high mobility and eventually slides outside the membrane. This case has been photocaptured and presented as a sequence of snapshots in Figure 4a. Initially, the droplet is impaled on the surface. Upon backpressure application, multiple gas pockets appear on the liquid−solid interface, deforming the droplet and gradually inciting downward movement. Not all of the gas pockets are of the same volume; some evolve faster than other, and some coalesce with each other. As seen in Figure 4a, the droplet is not completely levitated throughout the process. Of particular interest are the small openings at the contact line, as seen in Figure 4a(III), (IV). These are gas pockets residing near the contact line that break the contact line; with the gas flow, these openings are sustained or evolve up until the droplet moves. In some cases, mostly for large droplets and high backpressure values, small ejected droplets may appear, as in ref 66. Droplets with volume lower than 40 μL do not slide even at 90°, when no backpressure is applied. The inherently sticky surface gradually becomes slippery for a 40 μL droplet upon 0.11 bar backpressure application, exhibiting a sliding angle of

Figure 4. Water droplets upon backpressure application on a porous medium: (a) V = 60 μL and (b) V = 20 μL. For large droplets, multiple air pockets appear at the liquid−solid interface, whereas for smaller droplets one pocket finally evolves, leveraging the droplet sliding. In (b), smaller droplets are not moving, indicating that the actuation is volume-sensitive. (c) Schematic representation of the droplet actuation and mobility manipulation. (i) A quiescent droplet on an inclined surface, (ii) backpressure application and gas pocket appearance, resulting instantly in (iii) droplet actuation, (iv) droplet movement, and (v) the surface becoming sticky again when no backpressure is applied. C

DOI: 10.1021/acs.langmuir.6b00900 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

tions prohibit any strictly quantitative comparison with the experimental data, although we argue that the illuminated mechanisms of droplet movement are valid. Consider a water droplet with translational symmetry, of height ∼2 mm wetting a flat surface surrounded by air, inside a 2D computational domain of Ω = [6.2, 6.2] mm (Figure 6).

42°. At backpressure equal to 0.65 bar, the droplet exhibits high mobility, even at a horizontal porous surface. Similar behavior is recorded for the 20 and 10 μL droplets. For small droplets (V < 40 μL), one gas pocket is evolved in the liquid phase, thus providing adequate lubrication for the droplet to slide. However, this air cushion is not as fully developed as the one observed in Leidenfrost phenomena.61 This method for actuation and mobility manipulation may also be used for droplets of lower surface energy. In Figure 5,

Figure 6. Illustration of the 2D computational domain, Ω.

The governing equations are the following: (1) The momentum equations include an additional term that accounts for the surface tension force69

Figure 5. Sliding angle of 7.5% (mole fraction) ethanol in water droplets vs backpressure.

ρ

we present the sliding angle variation of a 7.5% ethanol (mole fraction) in water droplets vs the applied backpressure. This solution exhibits a surface tension of ca. 0.045 N/m, i.e., lower than water exhibiting 0.073 N/m. Again, the sliding angle decreases with the backpressure and with the droplet volume. However, in this case, the backpressure levels needed to render the surface slippery are lower compared to that for DI water. A 40 μL ethanol in water droplet becomes slippery in ca. 0.14 bar on a horizontal surface, whereas for a DI water droplet a backpressure of 0.65 bar was needed. Again, we did not observe a fully developed air cushion under the liquid. In all cases, for both large and small water droplets as well as for lower surface energy droplets, the mobility switching is reversible; i.e. upon backpressure relief, the surface returns to a sticky state and the droplet pins to it. After backpressure application, the droplet is again actuated as described before. Therefore, with this method porous surfaces act as active media for reversible switching between the sticky and slippery states.

∂u + ρ(u ·∇)u = −∇p + μ∇2 u + ρ g + G∇φ ∂t

(1)

where u is the flow velocity vector, p is the pressure, g is gravitational acceleration, G is the chemical potential (see below), and φ is the phase field, i.e., a spatial variable that indicates the present phase; φ = 1 indicates the presence of water, whereas φ = −1 indicates the presence of air. The density and viscosity depend on φ according to the relations 1+φ 1−φ 1+φ 1−φ ρ = 2 ρw + 2 ρa and μ = 2 μw + 2 μa , where ρa and μa are the density and viscosity of air and ρw and μw are the density and viscosity of water, respectively. (2) The Cahn− Hilliard phase-field equations are70 G=

λ ( −ε 2∇2 φ + φ3 − φ) ε2

∂φ + ∇·(φ u) = ∇·(M ∇G) ∂t

(2)

where λ is the mixing energy density that is related to the surface tension through σ = 2(2)1/2λ/(3ε), ε is the capillary width that controls the thickness of the diffuse interface, and M is the mobility of the diffuse interface. (3) The continuity equation, augmented with a source term that accounts for the air inlet at the pores, which are introduced to the model as point sources of mass, is

■

MECHANISM OF DROPLET ACTUATION AND MOVEMENT Depinning of the Receding Contact Line and Droplet Movement: Computational Approach. To provide insights into the mechanisms of droplet movement, we propose a model that simulates the dynamics of droplet actuation on porous surfaces. The model comprises the momentum conservation equations and the continuity equation along with the Cahn− Hilliard phase-field equations67,68 that provide the tracking of the involved phases (liquid−air). The current model is implemented in 2D as a result of the relatively low computational cost, where otherwise a full implementation in 3D would yield significantly higher computational costs, restricting more discerning analyses as presented below. The fundamental differences between the 2D and 3D implementa-

∂ρ + ∇·(ρ u) = ∂t

∑ Faδi i

(3)

where Fa is the mass flow rate at which air is entering the computational domain through each point source (pore) and δi is the Dirac delta function centered at the ith pore. The point sources are placed on the flat, solid surface, i.e., the bottom boundary of the computational domain, at a distance of 0.89 mm between each other, even though the pores are numerous and more densely distributed in reality. The compromise of D

DOI: 10.1021/acs.langmuir.6b00900 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

Figure 7. Simulation snapshots of droplet lateral movement on an inclined plane (notice the gravitational acceleration) with air permeating the solid substrate. The snapshots include the elapsed time, and they are arranged sequentially from left to right and top to bottom. The droplet profile is drawn against a spatially fixed vertical line that provides a reference for the droplet displacement; inclination = 30°, Fa = 2.45 × 10−5 kg/(s·m).

Figure 8. Comparison of snapshots from simulation and experiment, illustrating the mechanism of droplet movement. The collapse of the air bubble at the rear of the droplet sends a wave on the droplet surface that is traveling forward and can be seen in both simulations and experiments; the wave in the experiment snapshots is blurry because of the relatively slow camera shutter speed. The simulation snapshots correspond to a later stage of the simulation presented in Figure 7.

significance, given that most of the commonly used variables, e.g., pressure and velocity, fluctuate randomly as a result of the nature of two-phase flows. We opted for the mean velocity of the center of mass of the droplet (see below) that has statistical significance and physical importance. The refined mesh yielded only up to a 5% change in the mean velocity, which for the purposes of this work was deemed negligible; therefore, only the coarser mesh was systematically used. In Figure 7, we present simulation snapshots that outline a mechanism of droplet movement. The initial conditions of the simulations include a droplet on an inclined porous surface, not necessarily in an equilibrium state; i.e., the initial conditions are only estimations of the expected equilibrium wetting state. For t = 0 s, we enforce a certain rate at which air permeates the substrate (Fa); furthermore, we allow the effect of gravitational forces. Air pockets are gradually formed and expanded (Figure 7, t = 0.01 s), some of which coalesce into larger ones (Figure 7, t = 0.03 s). The droplet remains pinned until the stretched liquid bridge, formed between the outside air and the air pocket (Figure 7, t = 0.08 s) at the rear of the droplet (with respect to the direction of gravity), collapses. The collapse sends a wave forward on the surface of the droplet that reaches the solid surface on the front of the droplet (Figure 7, t = 0.086 to 0.094 s), at which point the droplet gets pinned on the surface (Figure 7, t = 0.096 s). Eventually, the droplet has skipped forward by the amount of the solid surface length that gets attached to the front of the droplet. This is the mainly observed mechanism of droplet movement that resembles droplet rolling, demonstrated also in Figure 8 (first row), for a later stage (t = 0.252−0.264 s). The collapse of the air bubble at the rear of the droplet and the wave on the droplet surface that is traveling forward can also be seen in the experiments as shown in Figure 8 (second row). The simulation results shown in Figures 7 and 8 refer to moderate values of Fa. If, however, Fa is reaches a certain limit, we observe levitation of the droplet on its larger part, while a

including fewer pores in the model is essential because otherwise it would require an extremely fine computational mesh that resolves all of the point sources. Furthermore, we expect that the inclusion of more pores will not have a considerable effect on the results because of the fast coalescence of air bubbles formed at the pores (presented below). The individual bubbles at each pore are generally shortlived and eventually coalesce into larger bubbles. Each established bubble is expanded by air entering through all of the assimilated pores, which have the equivalent effect of a single pore. This process is also suggested by experimental observations (Figure 4a), where the only observed bubbles are relatively large and countable, even though air is permeating the porous substrate through a significantly large number of pores. In the context of 2D simulations, the point sources (pores) represent infinitesimally narrow slits, extending in the direction of the translational symmetry plane. Therefore, the mass flow rate, Fa, is defined per unit length of the slits (kg/(s·m)). Contact angle hysteresis is modeled by periodically varying Young’s contact angle, θY, along the bottom boundary; θY is imposed as a boundary condition of eq 2.71 The value of θY follows a square waveform with amplitude 40° around the mean value of 90° and period of 0.31 mm. This approach yields realistic values of contact angle hysteresis of ∼20°, based on contact angle measurements. On the vertical boundaries of the computational domain, we impose periodic conditions in order to allow long traveling distances of the droplet while maintaining a small size of the domain. Finally, the domain, Ω, is tessellated into a mesh of ∼32 000 triangular elements, and eqs 1−3 are discretized with a finite element method, utilizing commercial software package COMSOL Multiphysics (ver. 4.3b). A denser mesh of ∼64 000 elements was also utilized in a mesh sensitivity study for selected values of Fa, namely, 2.45 × 10−5 and 6.13 × 10−5 kg/ (s·m). The criterion that satisfies the mesh sensitivity requirements has to be a calculated quantity with statistical E

DOI: 10.1021/acs.langmuir.6b00900 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

brief reversal of the movement mechanism, meaning that bubble collapse occurs at the front of the droplet (instead of collapse at the rear), sending a wave in the reverse direction. The reverse mechanism is rarely observed and ultimately has little impact on the mean velocity of the droplet, which has a monotonous increase with Fa (Figure 11), until the behavior of the system shifts to what was described above and is illustrated in Figure 9.

small part of the droplet is pinned to areas of the solid lacking pores (Figure 9). The pinning effectively binds the droplet to

Figure 9. Snapshot of a droplet (illustrated in gray color) pinned in the area between two pores, with its bulk suspended on a cushion of air entering through the pores. Streamlines are shown with black lines, and the velocity field magnitude (m/s) is illustrated by the color scale; t = 0.1 s, inclination = 30°, and Fa = 1.22 × 10−4 kg/(s·m).

Figure 11. Mean velocity (parallel component to the solid surface) of the droplet’s center of mass vs Fa. The regime denoted partial levitation has effectively zero mean velocity because the droplet is pinned. Beyond that regime, the droplet levitates completely and computations are no longer feasible (the droplet leaves the computational domain).

the surface, restricting any lateral movement, which is also illustrated by the velocity of the droplet’s center of mass (Figure 10, blue line). The velocity (component parallel to the

Calculation of the Total Pinned Fraction of the Macroscopic Contact Line. In most cases, the mobility of a droplet and the gradual conversion from a sticky to a slippery state are assessed by measuring the sliding angle (α), i.e., the minimum angle that a surface has to be inclined in order to instigate moving of the droplet. Many studies have been published on understanding the mechanisms of droplet movement.72−75 The governing equation is actually a balance of forces acting parallel to the surface, namely, the gravitational force and the pinning and hysteretic forces at the liquid/solid interface. At exactly α, i.e., at the verge of sliding or rolling, the balance takes the form ρgV sin α = k 2ωγ(cos θr − cos θa)

Figure 10. Velocity (parallel component to the solid surface) of the droplet’s center of mass vs time. Blue solid line, Fa = 1.22 × 10−4 kg/ (s·m); black solid line, Fa = 8.57 × 10−5 kg/(s·m). Dashed lines indicate the corresponding (color) mean velocities; inclination = 30°.

(4)

where ρ and γ are the density and surface tension of the liquid, respectively, V is the droplet volume, 2ω is the width of the drop in the direction perpendicular to inclination, and k is a retention-force factor that is expected to be equal to 1 but is usually adjusted to fit the experimental data. θr and θa are the receding and advancing contact angles, respectively, of the droplet on the respective smooth, solid surface. Discrepancies from this expression of force balance are far from rare as a result of different theoretical assumptions or different conditions for the experiments,75−78 whereas in this form it cannot explicitly account for the contributions of the capillary bridges developed in textured surfaces.51 A different approach in treating this problem is provided by assuming the in-plane retention forces75,79 or the vertical pinning forces at the receding contact line78 for each capillary bridge. In this latter approach, the vertical pinning force for each (j) capillary bridge is given by

surface) of the center of mass fluctuates mildly around 0; the fluctuations are caused by air entering from beneath, inciting droplet oscillations. This macroscopically static behavior has also been identified in the case of droplet motion using vertical vibration on an inclined surface.35,36 Further increases in Fa cause complete levitation of the droplet that eventually is ejected upward, at which point the computations break down because the droplet leaves the computational domain. For smaller values of Fa, the velocity fluctuates more aggressively in an almost irregular fashion (Figure 10, black line), having many instances with the center of mass traveling in the reverse direction (negative velocity), i.e., against gravity. The negative velocity is associated with a F

DOI: 10.1021/acs.langmuir.6b00900 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

Fj = Pjγ sin(θrj)

(5)

where Pj is the perimeter of an individual surface feature and θjr is the receding angle local to the jth feature. Contrary to the inplane pinning forces appearing in eq 4, this approach yields adhesion in the vertical direction

F /l = Φ0γ sin(θr)

(6)

where Φ0 is the total pinned fraction of a macroscopic contact line and can be considered to be a multiplier reflecting the strength of pinning on the textured surface compared to the respective smooth surface. l is the length of the macroscopic projected contact line. We are going to employ the model developed by Paxson et al.78 that also accounts for the vertical forces of each capillary bridge at the receding contact line, with appropriate modifications in order to account for the effect of the backpressure. FL is the force due to the Laplace pressure

⎛ 2γ ⎞ FL = πrb 2⎜ ⎟ ⎝ r ⎠

Figure 12. Schematic of the envisaged microscopic apparent contact line and capillary bridges atop the porous network.

variation of the pore size, FP may range from 6.08 × 10−10 to 6.23 × 10−9 N. Because we cannot measure the exact rb′ and θr′, we are going to calculate the total pinned fraction of the contact line after backpressure application, Φ′0, as the multiplier accounting for the decrease in FS to FS′ , encompassing the variations of rb and 2.6810−9

θr. Φ′0 = 4.9010−7 = 5.5 × 10−3, which is 2 to 3 orders of magnitude lower that that of the initial state. This value corroborates the mechanism revealed by the numerical approach presented before, entailing the partial depinning of the receding contact line, which triggers droplet actuation. The reduction of the pinned fraction is also supported experimentally (Figure 4a(III),(IV)). For larger droplets, the in-plane gravitational force is higher; therefore, a lower depinning fraction is needed to incite the downward movement, reflected by smaller backpressure values. The differences recorded for the case of ethanol in water cannot be attributed only to the effect of γ. The decrease in γ is ca. 40% (0.073 to 0.045 N/m), whereas the backpressure lowering is more than 77%, e.g., for a 50 μL droplet at zero inclination, 0.60 to 0.14 bar. The differences in this case stem from the different contact patch radius (rb) of the two liquids on the porous surfaces. Water and ethanol-in-water liquids are expected to exhibit different intrusion angles,80 thus corresponding to different rb values, which in turn is reflected in the vertical force balance.

(7)

where rb is the radius of the contact patch and r is the droplet radius. FS is the force due to the surface tension acting along the peripheral roughness features: FS = 2πrbΦ0γ sin(θr)

(8)

Φ0 is actually the fraction of the pinned contact line and can be considered to be a multiplier reflecting the strength of pinning on the textured surface compared to that on the respective smooth surface. In all cases, the gravitational force (FG) is given by FG = ρgV

(9)

The exact values of rb and Φ0 before and after backpressure application cannot be readily measured experimentally; therefore, assumptions will be drawn. On the basis of the SEM image given in Figure 1, the vast majority of rb ranges from ca. 0.50 to 1.65 μm. Because the particles do not exhibit additional scale roughness, it is reasonable to take Φ0 equal or close to 1. For a water droplet of 20 μL exhibiting r = 1.68 mm, well below the capillary length lc = γ /ρg ≈2.7 mm, before the backpressure application, for γ = 0.073 N/m, θr = 92°, rb = 1.07 μm, and Φ0 = 1, the pertinent forces are the following: FL = 3.12 × 10−10 N, and on the basis of the variation of rb, FL may range from 0.682 × 10−10 N to 7.34 × 10−10 N; FS = 4.90 × 10−7 N, and on the basis of the variation of rb, FS may range from 2.29 × 10−7 N to 7.51 × 10−7 N. Clearly, FL is 3 orders of magnitude smaller than FS; therefore, we are going to neglect it in the analysis hereafter. The gradual increase in backpressure, exerting a force FP on a quiescent droplet at a hydrophobic and inherently sticky incline, introduces manifold changes in the principle vertical force balance. A schematic of this process is given in Figure 12. With the application of the backpressure, FP is introduced into the force balance and the radius of the contact patch rb decreases to rb′ while θr decreases to θr′. For a 20 μL water droplet, the adequate pressure for sliding on a horizontal plane is 0.78 bar. On the basis of the SEM image in Figure 1, the pore radius ranges from ca. 0.5 to 1.6 μm. Taking the average, 1.05 μm, yields FP = PA = 2.68 × 10−9 N, with A being the pore opening area, and on the basis of the

■

CONCLUSIONS We provided insights into the underlying mechanisms of droplet actuation on porous media by means of backpressure and surface inclination; the latter are combined to facilitate active control of droplet mobility. With the adjustment of the porous backpressure, the droplet may be pinned to the surface or may be actuated and move downward by the presence of a small inclination. The interplay between backpressure and inclination has been quantified for various volumes of DI water and ethanol in DI water droplets. Actuation and mobility control may be realized without a fully developed air cushion under the liquid; therefore, Leidenfrost-like dynamics cannot be used. Simulations have shown that the droplet is actuated through a depinning process of the receding contact line and moves by means of forward wave propagation toward the front contact line. The droplet shapes obtained by this numerical approach are fairly compared to the ones observed experimentally, thus providing a reasonable justification of this approach. The small openings observed experimentally at the contact line during movement are followed by means of an analytical approach accounting for the vertical force balance and the total pinning fraction. G

DOI: 10.1021/acs.langmuir.6b00900 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

■

(14) Gogolides, E.; Vlachopoulou, M.; Tsougeni, K.; Vourdas, N.; Tserepi, A. Micro and nano structuring and texturing of polymers using plasma processes: potential manufacturing applications. Int. J. Nanomanuf. 2010, 6 (1), 152−163. (15) Schutzius, T. M.; Jung, S.; Maitra, T.; Eberle, P.; Antonini, C.; Stamatopoulos, C.; Poulikakos, D. Physics of Icing and Rational Design of Surfaces with Extraordinary Icephobicity. Langmuir 2015, 31 (17), 4807−4821. (16) Davis, A.; Yeong, Y. H.; Steele, A.; Bayer, I. S.; Loth, E. Superhydrophobic Nanocomposite Surface Topography and Ice Adhesion. ACS Appl. Mater. Interfaces 2014, 6 (12), 9272−9279. (17) Mosadegh-Sedghi, S.; Rodrigue, D.; Brisson, J.; Iliuta, M. C. Wetting phenomenon in membrane contactors − Causes and prevention. J. Membr. Sci. 2014, 452, 332−353. (18) Cheah, M. J.; Kevrekidis, I. G.; Benziger, J. B. Water slug formation and motion in gas flow channels: the effects of geometry, surface wettability, and gravity. Langmuir 2013, 29 (31), 9918−9934. (19) Zhang, F. Y.; Yang, X. G.; Wang, C. Y. Liquid water removal from a polymer electrolyte fuel cell. J. Electrochem. Soc. 2006, 153 (2), A225−A232. (20) Ghosh, A.; Beaini, S.; Zhang, B. J.; Ganguly, R.; Megaridis, C. M. Enhancing Dropwise Condensation through Bioinspired Wettability Patterning. Langmuir 2014, 30 (43), 13103−13115. (21) Attinger, D.; Frankiewicz, C.; Betz, A. R.; Schutzius, T. M.; Ganguly, R.; Das, A.; Kim, C.-J.; Megaridis, C. M. Surface engineering for phase change heat transfer: A review. MRS Energy & Sustainability 2014, 1, E4. (22) Khalil, K. S.; Mahmoudi, S. R.; Abu-dheir, N.; Varanasi, K. K. Active surfaces: Ferrofluid-impregnated surfaces for active manipulation of droplets. Appl. Phys. Lett. 2014, 105 (4), 041604. (23) Zhu, Y.; Antao, D. S.; Xiao, R.; Wang, E. N. Tunable Surfaces: Real-Time Manipulation with Magnetically Tunable Structures (Adv. Mater. 37/2014). Adv. Mater. 2014, 26 (37), 6357−6357. (24) Yao, X.; Hu, Y.; Grinthal, A.; Wong, T. S.; Mahadevan, L.; Aizenberg, J. Adaptive fluid-infused porous films with tunable transparency and wettability. Nat. Mater. 2013, 12, 529−534. (25) Stamatopoulos, C.; Schutzius, T. M.; Köppl, C. J.; El Hayek, N.; Maitra, T.; Hemrle, J.; Poulikakos, D. On the shedding of impaled droplets: The role of transient intervening layers. Sci. Rep. 2016, 6, 18875. (26) Verplanck, N.; Coffinier, Y.; Thomy, V.; Boukherroub, R. Wettability Switching Techniques on Superhydrophobic Surfaces. Nanoscale Res. Lett. 2007, 2 (12), 577−596. (27) Reyssat, M.; Richard, D.; Clanet, C.; Quere, D. Dynamical superhydrophobicity. Faraday Discuss. 2010, 146, 19−33. (28) Whyman, G.; Bormashenko, E. Wetting transitions on rough substrates: General considerations. J. Adhes. Sci. Technol. 2012, 26, 207−220. (29) Bormashenko, E.; Musin, A.; Whyman, G.; Zinigrad, M. Wetting Transitions and Depinning of the Triple Line. Langmuir 2012, 28 (7), 3460−3464. (30) Bormashenko, E. Progress in understanding wetting transitions on rough surfaces. Adv. Colloid Interface Sci. 2015, 222, 92−103. (31) Krupenkin, T. N.; Taylor, J. A.; Wang, E. N.; Kolodner, P.; Hodes, M.; Salamon, T. R. Reversible Wetting−Dewetting Transitions on Electrically Tunable Superhydrophobic Nanostructured Surfaces. Langmuir 2007, 23 (18), 9128−9133. (32) Zhang, J.; Feng, H.; Zao, W.; Zhao, Y.; Zhang, H.; Liu, Y. Magnetically driven superhydrophobic meshes with the capacity of moving at air/water and oil/water interfaces. RSC Adv. 2015, 5 (59), 47892−47899. (33) Sammarco, T. S.; Burns, M. A. Thermocapillary pumping of discrete drops in microfabricated analysis devices. AIChE J. 1999, 45 (2), 350−366. (34) Boreyko, J. B.; Collier, C. P. Dewetting Transitions on Superhydrophobic Surfaces: When Are Wenzel Drops Reversible? J. Phys. Chem. C 2013, 117, 18084.

AUTHOR INFORMATION

Corresponding Authors

*E-mail: N. Vourdas, [email protected]. *E-mail: V. N. Stathopoulos, [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS N.V. and V.N.S. have received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement no. 680599 (project title: Industrial Thermal Energy Recovery Conversion and Management). G.P., G.K., and A.G.B. have received funding from the Hellenic National Strategic Reference Framework (NSRF) 2007−2013 under grant agreement no. MIS380835 (project title: THALES − DESIREDROP: DESIgn and fabrication of Robust supErhyDROPhobic/philic surfaces and their application in the realization of “smart” microfluidic valves).

■

REFERENCES

(1) de Gennes, P. G.; Brochard-Wyart, F.; Quere, D. Capillarity and Wetting Phenomena: Drops, Bubbles, Pearls, Waves; Spinger-Verlag: New York, 2004. (2) Glavan, A. C.; Martinez, R. V.; Maxwell, E. J.; Subramaniam, A. B.; Nunes, R. M.; Soh, S.; Whitesides, G. M. Rapid fabrication of pressure-driven open-channel microfluidic devices in omniphobic R(F) paper. Lab Chip 2013, 13 (15), 2922−30. (3) Vourdas, N.; Tserepi, A.; Boudouvis, A.; Gogolides, E. Plasma processing for polymeric microfluidics fabrication and surface modification: Effect of super-hydrophobic walls on electroosmotic flow. Microelectron. Eng. 2008, 85 (5), 1124−1127. (4) Moschou, D.; Vourdas, N.; Kokkoris, G.; Papadakis, G.; Parthenios, J.; Chatzandroulis, S.; Tserepi, A. All-plastic, low-power, disposable, continuous-flow PCR chip with integrated microheaters for rapid DNA amplification. Sens. Actuators, B 2014, 199, 470−478. (5) Kitsara, M.; Ducrée, J. Integration of functional materials and surface modification for polymeric microfluidic systems. J. Micromech. Microeng. 2013, 23 (3), 033001. (6) Ghosh, A.; Ganguly, R.; Schutzius, T. M.; Megaridis, C. M. Wettability patterning for high-rate, pumpless fluid transport on open, non-planar microfluidic platforms. Lab Chip 2014, 14 (9), 1538−1550. (7) Abdelgawad, M.; Freire, S. L. S.; Yang, H.; Wheeler, A. R. Allterrain droplet actuation. Lab Chip 2008, 8 (5), 672−677. (8) Gogolides, E.; Ellinas, K.; Tserepi, A. Hierarchical micro and nano structured, hydrophilic, superhydrophobic and superoleophobic surfaces incorporated in microfluidics, microarrays and lab on chip microsystems. Microelectron. Eng. 2015, 132, 135−155. (9) Song, H.; Chen, D. L.; Ismagilov, R. F. Reactions in droplets in microfluidic channels. Angew. Chem., Int. Ed. 2006, 45 (44), 7336−56. (10) Abdelaziz, R.; Disci-Zayed, D.; Hedayati, M. K.; Pöhls, J.-H.; Zillohu, A. U.; Erkartal, B.; Chakravadhanula, V. S. K.; Duppel, V.; Kienle, L.; Elbahri, M. Green chemistry and nanofabrication in a levitated Leidenfrost drop. Nat. Commun. 2013, 4, 2400. (11) Vourdas, N.; Tserepi, A.; Gogolides, E. Nanotextured superhydrophobic transparent poly (methyl methacrylate) surfaces using high-density plasma processing. Nanotechnology 2007, 18 (12), 125304. (12) Vourdas, N. E.; Vlachopoulou, M. E.; Tserepi, A.; Gogolides, E. Nano-textured polymer surfaces with controlled wetting and optical properties using plasma processing. Int. J. Nanotechnol. 2009, 6 (1−2), 196−207. (13) Tsougeni, K.; Vourdas, N.; Tserepi, A.; Gogolides, E.; Cardinaud, C. Mechanisms of oxygen plasma nanotexturing of organic polymer surfaces: from stable super hydrophilic to super hydrophobic surfaces. Langmuir 2009, 25 (19), 11748−11759. H

DOI: 10.1021/acs.langmuir.6b00900 Langmuir XXXX, XXX, XXX−XXX

Article

Langmuir

(57) Vourdas, N.; Stathopoulos, V. N. Droplet Actuation and Droplet Mobility Manipulation on Porous Surfaces by Means of Backpressure Control. Technical Proceedings of the 2014 NSTI Nanotechnology Conference and Expo, NSTI-Nanotech 2014; Vol. 1, pp 220−223. (58) Stathopoulos, V. N.; Vourdas, N. Reversible wetting transitions on porous materials. The 8th International Conference on Material Technologies and Modeling, MMT 2014; pp 4−144. (59) Vourdas, N.; Dalamagkidis, K.; Stathopoulos, V. Active porous valves for plug actuation and plug flow manipulation in open channel fluidics. RSC Adv. 2015, 5 (126), 104594−104600. (60) Vourdas, N.; Dalamagkidis, K.; Stathopoulos, V. N. uVALVIT: A tool for droplet mobility control and valving. MATEC Web Conf. 2016, 41.0400310.1051/matecconf/20164104003 (61) Quéré, D. Leidenfrost Dynamics. Annu. Rev. Fluid Mech. 2013, 45 (1), 197−215. (62) Snoeijer, J.; Brunet, P.; Eggers, J. Maximum size of drops levitated by an air cushion. Phys. Rev. E 2009, 79 (3), 036307. (63) Snoeijer, J. H.; van der Weele, K. Physics of the granite sphere fountain. Am. J. Phys. 2014, 82 (11), 1029−1039. (64) Vourdas, N.; Dalamagkidis, K.; Kostis, I.; Vasilopoulou, M.; Davazoglou, D. Omnidirectional antireflective properties of porous tungsten oxide films with in-depth variation of void fraction and stoichiometry. Opt. Commun. 2012, 285 (24), 5229−5234. (65) Kostis, I.; Vasilopoulou, M.; Soultati, A.; Argitis, P.; Konofaos, N.; Douvas, A. M.; Vourdas, N.; Papadimitropoulos, G.; Davazoglou, D. Highly porous tungsten oxides for electrochromic applications. Microelectron. Eng. 2013, 111, 149−153. (66) Tan, J. N.; Ma, C.; Sivanantha, N.; Neild, A. Bubble inducing cell lysis in a sessile droplet. Appl. Phys. Lett. 2014, 104 (10), 103704. (67) Pashos, G.; Kokkoris, G.; Boudouvis, A. G. Minimum energy paths of wetting transitions on grooved surfaces. Langmuir 2015, 31 (10), 3059−3068. (68) Pashos, G.; Kokkoris, G.; Boudouvis, A. G. A modified phasefield method for the investigation of wetting transitions of droplets on patterned surfaces. J. Comput. Phys. 2015, 283, 258−270. (69) Yue, P.; Zhou, C.; Feng, J. J.; Ollivier-Gooch, C. F.; Hu, H. H. Phase-field simulations of interfacial dynamics in viscoelastic fluids using finite elements with adaptive meshing. J. Comput. Phys. 2006, 219 (1), 47−67. (70) Jacqmin, D. Calculation of two-phase Navier-Stokes flows using phase-field modeling. J. Comput. Phys. 1999, 155 (1), 96−127. (71) Lee, H. G.; Kim, J. Accurate contact angle boundary conditions for the Cahn-Hilliard equations. Comput. Fluids 2011, 44 (1), 178− 186. (72) Furmidge, C. G. L. Studies at phase interfaces. I. The sliding of liquid drops on solid surfaces and a theory for spray retention. J. Colloid Sci. 1962, 17 (4), 309−324. (73) Quéré, D.; Azzopardi, M.-J.; Delattre, L. Drops at Rest on a Tilted Plane. Langmuir 1998, 14 (8), 2213−2216. (74) Krasovitski, B.; Marmur, A. Drops down the hill: Theoretical Study of Limiting Contact Angles and the Hysteresis Range on a Tilted Plate. Langmuir 2005, 21, 3881−3885. (75) Extrand, C. W.; Moon, S. I. Repellency of the Lotus Leaf: Contact Angles, Drop Retention, and Sliding Angles. Langmuir 2014, 30 (29), 8791−8797. (76) Santos, M. J.; Velasco, S.; White, J. A. Simulation Analysis of Contact Angles and Retention Forces of Liquid Drops on Inclined Surfaces. Langmuir 2012, 28 (32), 11819−11826. (77) Smyth, K. M.; Varanasi, K. K.; Kwon, H.-M.; Paxson, A. T. Visualization of contact line motion on hydrophobic textures. Surf. Innovations 2013, 1 (2), 84−91. (78) Paxson, A. T.; Varanasi, K. K. Self-similarity of contact line depinning from textured surfaces. Nat. Commun. 2013, 4, 1492. (79) Pilat, D. W.; Papadopoulos, P.; Schaffel, D.; Vollmer, D.; Berger, R.; Butt, H. J. Dynamic measurement of the force required to move a liquid drop on a solid surface. Langmuir 2012, 28 (49), 16812−20. (80) Extrand, C. W.; Moon, S. I. Intrusion pressure to initiate flow through pores between spheres. Langmuir 2012, 28 (7), 3503−3509.

(35) Sartori, P.; Quagliati, D.; Varagnolo, S.; Pierno, M.; Mistura, G.; Magaletti, F.; Casciola, C. M. Drop motion induced by vertical vibrations. New J. Phys. 2015, 17 (11), 113017. (36) Brunet, P.; Eggers, J.; Deegan, R. D. Vibration-induced climbing of drops. Phys. Rev. Lett. 2007, 99 (14), 144501. (37) Lee, J.; He, B.; Patankar, N. A. A roughness-based wettability switching membrane device for hydrophobic surfaces. J. Micromech. Microeng. 2005, 15 (3), 591. (38) Park, S.-Y.; Chiou, P.-Y. Light-driven droplet manipulation technologies for lab-on-a-chip applications. Adv. OptoElectron. 2011, 2011, 909174. (39) Jin, M.; Li, S.; Wang, J.; Xue, Z.; Liao, M.; Wang, S. Underwater superoleophilicity to superoleophobicity: role of trapped air. Chem. Commun. 2012, 48 (96), 11745−11747. (40) de Ruiter, J.; Lagraauw, R.; van den Ende, D.; Mugele, F. Wettability-independent bouncing on flat surfaces mediated by thin air films. Nat. Phys. 2014, 11 (1), 48−53. (41) Vourdas, N.; Tserepi, A.; Stathopoulos, V. N. Reversible pressure-induced switching of droplet mobility after impingement on porous surface media. Appl. Phys. Lett. 2013, 103 (11), 111602. (42) Bouwhuis, W.; Hendrix, M. H. W.; van der Meer, D.; Snoeijer, J. H. Initial surface deformations during impact on a liquid pool. J. Fluid Mech. 2015, 771, 503−519. (43) Schutzius, T. M.; Jung, S.; Maitra, T.; Graeber, G.; Köhme, M.; Poulikakos, D. Spontaneous droplet trampolining on rigid superhydrophobic surfaces. Nature 2015, 527 (7576), 82−85. (44) Lee, C.; Kim, C.-J. Underwater Restoration and Retention of Gases on Superhydrophobic Surfaces for Drag Reduction. Phys. Rev. Lett. 2011, 106 (1), 014502. (45) Verho, T.; Korhonen, J. T.; Sainiemi, L.; Jokinen, V.; Bower, C.; Franze, K.; Franssila, S.; Andrew, P.; Ikkala, O.; Ras, R. H. A. Reversible switching between superhydrophobic states on a hierarchically structured surface. Proc. Natl. Acad. Sci. U. S. A. 2012, 109 (26), 10210−10213. (46) Tan, J. N.; Ling, W. Y. L.; Neild, A. Selective Liquid Droplet Transfer Using Injected Bubbles. Appl. Phys. Express 2013, 6 (7), 077301. (47) Karatay, E.; Haase, A. S.; Visser, C. W.; Sun, C.; Lohse, D.; Tsai, P. A.; Lammertink, R. G. Control of slippage with tunable bubble mattresses. Proc. Natl. Acad. Sci. U. S. A. 2013, 110 (21), 8422−6. (48) Antonini, C.; Carmona, F. J.; Pierce, E.; Marengo, M.; Amirfazli, A. General methodology for evaluating the adhesion force of drops and bubbles on solid surfaces. Langmuir 2009, 25 (11), 6143−54. (49) Ling, W. Y.; Lu, G.; Ng, T. W. Increased stability and size of a bubble on a superhydrophobic surface. Langmuir 2011, 27 (7), 3233− 7. (50) Ling, W. Y.; Ng, T. W.; Neild, A. Effect of an encapsulated bubble in inhibiting droplet sliding. Langmuir 2010, 26 (22), 17695− 702. (51) Liang Ling, W. Y.; Ng, T. W.; Neild, A.; Zheng, Q. Sliding variability of droplets on a hydrophobic incline due to surface entrained air bubbles. J. Colloid Interface Sci. 2011, 354 (2), 832−42. (52) Goldshtik, M. A.; Khanin, V. M.; Ligai, V. G. A liquid drop on an air cushion as an analogue of Leidenfrost boiling. J. Fluid Mech. 1986, 166, 1−20. (53) Brunet, P.; Snoeijer, J. H. Star-drops formed by periodic excitation and on an air cushion − A short review. Eur. Phys. J.: Spec. Top. 2011, 192 (1), 207−226. (54) Bouwhuis, W.; Winkels, K. G.; Peters, I. R.; Brunet, P.; van der Meer, D.; Snoeijer, J. H. Oscillating and star-shaped drops levitated by an airflow. Phys. Rev. E 2013, 88 (2), 023017. (55) Papoular, M.; Parayre, C. Gas-film levitated liquids: Shape fluctuations of viscous drops. Phys. Rev. Lett. 1997, 78 (11), 2120− 2123. (56) Vourdas, N.; Ranos, C.; Stathopoulos, V. N. Reversible and dynamic transitions between sticky and slippery states on porous surfaces with ultra-low backpressure. RSC Adv. 2015, 5 (42), 33666− 33673. I

DOI: 10.1021/acs.langmuir.6b00900 Langmuir XXXX, XXX, XXX−XXX